Reinforcement learning for playing games

Project Level: Winter

Project Duration:

4 weeks – 20-36 hours per week. Applicant will be required on-site for the project.

Description:

Have you played a game with an AI before? Do you wonder how AI can play games well? Recently reinforcement learning (RL) has been used to
build super-human AI for playing many games. This includes the AlphaGo algorithm which clinched sensational victories over top human players in
Go. The successful applicant of this project will implement state-of-the-art RL-based AI for playing several popular games, and will have the opportunity to develop new algorithms if time permits.

Expected Outcomes:

Gain knowledge on RL and some state-of-the art RL algorithms.
Develop the ability to implement RL-based AI for playing games.
Develop skills in research design, implementation, experimentation, and communication.
A report documenting the work done and the findings.

Suitable for:

An applicant should have good knowledge on deep learning and good programming skills.

Further Information:

Email Dr Nan Ye for any inquiry on the project.